William J. Gedney's elicitation questionnaire.

• Author: Hudak, Thomas John

William J. Gedney was well known for the accurate and extensive data that he collected in his work on Tai dialects and languages. Unlike some investigators who actually settled in a village to learn and then record a dialect, Gedney worked methodically with a large number of individual informants to elicit data. After his retirement in 1980, he often spoke of publishing the questionnaire he had used for the benefit of other researchers. Part of this questionnaire, that for determining tonal systems, appeared in his 1972 article, "A Checklist for Determining Tones in Tai Dialects." But it was only after his death in 1999 that his original notebooks with the remaining portion of the questionnaire emerged. That questionnaire, along with the one for tonal systems, is presented here.

The Tai family of languages extends across Southeast Asia from Assam in the west to the island of Hainan in the east and from Yunnan, Guangxi, and Guangdong in southern China in the north to the Thai-Malay border in the south. Ahom, now an extinct language, was found in Assam, the farthest point west where the Tai languages are spoken. Shan and related dialects occur in Burma and Thailand as well as along the border in Assam and Yunnan. Lue speakers inhabit the Sipsongpanna region in southern Yunnan and in western areas of Laos. In Thailand Siamese (Thai) and closely related dialects cover the area, while Lao and Lao dialects are spoken in northeastern Thailand and Laos. White, Black, and Red Tai along with Yay are spoken in northwest Vietnam, sometimes spilling into Laos, while Tay (Tho) and Nung can be found in northeast Vietnam. In China, Zhuang speakers can be found throughout Guangxi and in border areas of Yunnan and Guangdong. Throughout these areas intermediate varieties of Tai languages also occur, usually identified by place names. And at the farthest point east, in Hainan, there exist a number of languages that appear to be closely related to the Tai family. Following Fang-kuei Li, scholars generally divide this family into three branches: the Northern, the Central, and the Southwestern.

Research on these languages and the Tai family has been in progress for many years, with some exceptional work done in the nineteenth century in the form of Siamese and Shan dictionaries. However, by and large, this early work was done by individuals with little or no linguistic training. Transcriptions, and the resulting publications, were impressionistic, filled with errors and inaccuracies. Often tones were disregarded, vowel length undifferentiated, and unfamiliar sounds ignored. Only in the twentieth century with the work of Fang-kuei Li, Andre-G. Haudricourt, and William J. Gedney was the rigorous and scientific investigation of these languages initiated.

William J. Gedney's investigation of the Tai languages began with his studies of Siamese in the 1940s and 1950s and then with other Tai languages beginning in 1964. Noted for the great care with which he recorded tonal and phonological systems, Gedney described the development of his approach in the following manner. Before going to the field in 1964, he examined previous word lists used by scholars and selected those items which had appeared in two or more of the Tai languages. To these he added words that he felt were inherited Tai words and that would appear in other Tai languages. These words were then arranged in sets such that all the words that he suspected would have the same tone would appear in one set. For example, those words that were thought to have a rising tone would appear in one set, those with a low tone in another, and so on through all of the tone categories. Those that diverged from the expected tone would be readily apparent. Such an arrangement, however, proved to be impractical, causing confusion for the informant as the questioning jumped from one semantic field to another. Gedney then divided his approach. He first selected a small group of words from the list to determine the tonal system of the dialect under investigation. The remaining words in the word list were then divided into groups based on cultural and semantic categories. Before any elicitation Gedney meticulously recorded the name of the informant, the language recorded, the date, and the place of the elicitation.

TEST WORDS FOR TONES

All scholars hold that the Tai languages had their origin in an earlier system known as Proto-Tai that had three tones on open syllables and no tonal contrast on checked syllables. Tones on open syllables have been conventionally designated as A, B, C, and those on checked syllables as D. In each of these categories, the tones underwent phonemic splits, conditioned by the phonetic nature of the initial consonants. The first split depended upon whether the initial consonants were voiced or voiceless. This binary split allowed for a maximum of six different tones on the open syllables. For the D category, the splits were conditioned not only by the voiced-voiceless split but also by vowel length. The voiceless group had further splits dependent on 1) groups of friction sounds such as voiceless fricatives (s, f, x, h), voiceless aspirated stops (ph, th, kh), and preaspirated sonorants (hm, hm, hl); 2) voiceless unaspirated sounds (p, t, k); and 3) glottal and preglottalized sounds (?, ?b, ?d). All of these splits would allow for a possible maximum of twenty contrasts as shown in the following chart:

 Proto-tai Tones A B C D-short D-long voiceless friction sounds, *s, hm, etc. 1 5 9 13 17 voiceless unaspirated sounds, *p, t, k 2 6 10 14 18 glottal sounds, *?,?b, etc. 3 7 11 15 19 voiced sounds, *b, m, l, etc. 4 8 12 16 20 

These splits are explained in greater detail in Gedney 1972.

From the original questionnaire Gedney created the following list to determine the tonal system. More than one word was selected for each possible box in the event that a particular dialect did not have the sample word or that it had a different or non-Tai word. Gedney (1972: 429) comments regarding this list, "Test words are given with their Siamese consonant and vowel values (or those they would have if they occurred in Siamese, as in the case of the word for 'arrow of crossbow'), but without indication of tone, since it is precisely tones that are being sought."

[TABLE OMITTED]

[TABLE OMITTED]

SEMANTIC AND CULTURAL BASED QUESTIONNAIRE

Having uncovered the tonal system, Gedney then turned to the following list arranged by semantic and cultural categories. The list consists of words that the Tai languages are known to have and that a comparativist needs to know. Occasionally a word completely unrelated to the sphere is requested, simply because it is convenient. For example, the word for 'crucible' is elicited after the term 'eye socket' to determine whether or not baw (3) is common to both of them. In most cases one set of words leads naturally into another and informants on their own initiative will add additional related words to the sphere. During his interviews Gedney used Siamese or some mutually intelligible language such as Lao or Black Tai, not English. In the notebooks the Siamese words are presented in list form with tones indicated by diacritics rather than superscript numbers. Here the superscripts indicate the following Siamese tones: 1-mid, 2-low, 3-falling, 4-high, 5-rising. In a later unpublished manuscript (n.d.), Gedney merely presented the list in English with subgroups within each cultural sphere separated by semicolons. For greater clarity here, the first word in a subgroup is capitalized.

1. Numbers. One ([TEXT NOT REPRODUCIBLE IN ASCII]), one, single ([TEXT NOT REPRODUCIBLE IN ASCII]), same ([TEXT NOT REPRODUCIBLE IN ASCII]), two ([TEXT NOT REPRODUCIBLE IN ASCII]), three ([TEXT NOT REPRODUCIBLE IN ASCII]), four ([TEXT NOT REPRODUCIBLE IN ASCII]), five ([TEXT NOT REPRODUCIBLE IN ASCII]), six ([TEXT NOT REPRODUCIBLE IN ASCII]), seven ([TEXT NOT REPRODUCIBLE IN ASCII]), eight ([TEXT NOT REPRODUCIBLE IN ASCII]), nine ([TEXT NOT REPRODUCIBLE IN ASCII]), ten ([TEXT NOT REPRODUCIBLE IN ASCII]), eleven ([TEXT NOT REPRODUCIBLE IN ASCII]), twelve ([TEXT NOT REPRODUCIBLE IN ASCII]); Twenty ([TEXT NOT REPRODUCIBLE IN ASCII]), thirty ([TEXT NOT REPRODUCIBLE IN ASCII]); Hundred ([TEXT NOT REPRODUCIBLE IN ASCII]), 101 or 110 ([TEXT NOT REPRODUCIBLE IN ASCII]), 110 ([TEXT NOT REPRODUCIBLE IN ASCII]), thousand ([TEXT NOT REPRODUCIBLE IN ASCII]), ten thousand ([TEXT NOT REPRODUCIBLE IN ASCII]), hundred thousand ([TEXT NOT REPRODUCIBLE IN ASCII]), million ([TEXT NOT REPRODUCIBLE IN ASCII]); Even ([TEXT NOT REPRODUCIBLE IN ASCII]), odd ([TEXT NOT REPRODUCIBLE IN ASCII]).

2. Head and Hair. Hair of the head ([TEXT NOT REPRODUCIBLE IN ASCII]), grey hair ([TEXT NOT REPRODUCIBLE IN ASCII]), whorl in the hair ([TEXT NOT REPRODUCIBLE IN ASCII]), same as 'fur in animals'? 'spirit'?), topknot ([TEXT NOT REPRODUCIBLE IN ASCII]), woman's hairknot ([TEXT NOT REPRODUCIBLE IN ASCII]); False hair ([TEXT NOT REPRODUCIBLE IN ASCII]); To cut the hair ([TEXT NOT REPRODUCIBLE IN ASCII]), to wash the hair ([TEXT NOT REPRODUCIBLE IN ASCII]), to shave the head ([TEXT NOT REPRODUCIBLE IN ASCII]); Short hair ([TEXT NOT REPRODUCIBLE IN ASCII]), long hair ([TEXT NOT REPRODUCIBLE IN ASCII]); The hair falls out ([TEXT NOT REPRODUCIBLE IN ASCII]); A comb ([TEXT NOT REPRODUCIBLE IN ASCII]), to comb ([TEXT NOT REPRODUCIBLE IN ASCII]).

3. Head and Face. Back of the neck ([TEXT NOT REPRODUCIBLE IN ASCII]), face ([TEXT NOT REPRODUCIBLE IN ASCII]), forehead ([TEXT NOT REPRODUCIBLE IN ASCII]), cheek ([TEXT NOT REPRODUCIBLE IN ASCII]), pale ([TEXT NOT REPRODUCIBLE IN ASCII]), chin ([TEXT NOT REPRODUCIBLE IN ASCII]), sideburns ([TEXT NOT REPRODUCIBLE IN ASCII]), temples ([TEXT NOT REPRODUCIBLE IN ASCII]), fontanelle ([TEXT NOT REPRODUCIBLE IN ASCII]).

4. The Eyes. Eyebrow ([TEXT NOT REPRODUCIBLE IN ASCII]), eyelash ([TEXT NOT REPRODUCIBLE IN ASCII]), eyelid ([TEXT NOT REPRODUCIBLE IN ASCII]), upper eyelid ([TEXT...